Carillion — dash-for-cash contracting collapse

1. Episode summary

Carillion plc was a UK-listed construction and facilities-management group demerged from Tarmac in 1999. Over the following decade it assembled its scale through debt-financed acquisitions, most importantly Mowlem (£350m, 2006), Alfred McAlpine (£572m, 2008) and Eaga (£298m, 2011), absorbing sizeable inherited pension deficits in the process. By 2016 the group employed around 43,000 people, held roughly 420 UK public-sector contracts, and had increased its dividend in each of the sixteen years since formation. Its operating model combined thin-margin construction work — including a Battersea Power Station phase bid at a 0% profit margin — with longer-dated Private Finance Initiative and facilities-management contracts, and relied heavily on reverse-factoring "supply-chain finance" to lengthen payables. Between 2012 and 2016 the pension buyout deficit widened toward £2.6bn while dividends and executive bonuses continued to be paid. On 10 July 2017 Carillion announced an £845m provision, concentrated on three PFI construction schemes (Royal Liverpool Hospital, Midland Metropolitan Hospital, Aberdeen Western Peripheral Route); its chief executive resigned. Further writedowns, a breached covenant and failed rescue talks with government and lenders followed. On 15 January 2018 Carillion entered compulsory liquidation with liabilities of roughly £7bn, triggering the largest UK trading-liquidation to date, taxpayer costs estimated by the National Audit Office at around £148m, and loss of pension entitlements for tens of thousands of scheme members. The strategic question the episode turned on was whether directors and auditors could continue to recognise contract revenue, pay dividends and acquire businesses on the basis of optimistic traffic-light assessments that did not reflect the cash and risk profile of the underlying contract book.

2. Sources

Primary:

1. House of Commons Business, Energy and Industrial Strategy Committee & Work and Pensions Committee, Carillion — Second Joint Report of Session 2017–19, HC 769, 16 May 2018 (full report and summary).
2. National Audit Office, Investigation into the government's handling of the collapse of Carillion, HC 1002, June 2018 (£148m taxpayer-cost estimate; timeline of Cabinet Office contingency planning).
3. National Audit Office, Investigation into the rescue of Carillion's PFI hospital contracts, HC 14, January 2020 (loss breakdowns on Midland Metropolitan and Royal Liverpool PFI schemes; ≈£603m combined loss to shareholders, lenders, insurers and Carillion on the two hospitals).
4. The Pensions Regulator, The Carillion Group — Regulatory Intervention Report, 2020 (trustees' 2010 and 2013 requests for formal intervention; £2.6bn buyout deficit on collapse).
5. Carillion plc, Trading update and contract review announcement, 10 July 2017 (London Stock Exchange RNS), recording the £845m provision, withdrawal of guidance, suspension of dividend, and CEO resignation.
6. Financial Reporting Council, Sanctions against KPMG LLP, KPMG Audit plc and two former partners — Carillion, 12 October 2023 (and 2022 tribunal decision on AQR dishonesty findings): record £30m gross fine on KPMG, partner sanctions relating to the 2014–2016 Carillion audits.
7. Financial Conduct Authority, Final Notice — Richard Adam, FCA Ref, issued January 2026: £232,800 fine; findings that Adam, as Finance Director April 2007–December 2016, bore responsibility for procedures, systems and controls relating to financial reporting that were insufficient to ensure contract-accounting judgements in UK construction were made, recorded and reported appropriately; recklessness finding under Market Abuse Regulation and Listing Rules.
8. Financial Conduct Authority, Final Notice — Richard Howson, FCA Ref, issued January 2026: £237,700 fine; findings that Howson as Chief Executive was aware of serious financial troubles in Carillion's UK construction business but failed to reflect this in company announcements or alert the Board and audit committee; recklessness finding under Market Abuse Regulation and Listing Rules.
9. Financial Conduct Authority, Final Notice — Zafar Khan, FCA Ref, issued January 2026: £138,900 fine; findings relating to Khan's role as Finance Director (successor to Adam) in misleading investor statements in the period up to the July 2017 profit warning; recklessness finding under Market Abuse Regulation and Listing Rules.

Secondary (with justification):

1. House of Commons Library, The collapse of Carillion, Briefing Paper CBP-8206, March 2018 — parliamentary research service synthesis drawing on committee evidence and regulator statements.
2. Institute for Government, Carillion: two years on, January 2020 — policy-research review analysing the outsourcing and PFI context of the collapse.
3. London School of Economics Law Review, "The Collapse of Carillion: Regulatory Failure in the Contract State", Vol. 4 (2019) — peer-reviewed legal/regulatory analysis of contract accounting, audit and oversight failures.
4. ACCA, Carillion's collapse and the future of PFI, 2018 — professional-body technical analysis of the PFI contracting and reverse-factoring mechanics.
5. London Business School, "Two lessons from the failure of Carillion" (think-piece drawing on committee evidence) — management-school analytical synthesis.
6. Duke University School of Law, FinReg Blog, "Carillion plc: A Governance Case Study from the UK", 18 July 2018 — draws on HC 769 and Carillion annual reports; synthesises board composition (seven-member board: CEO Howson, FD Khan, Chairman Philip Green from May 2014, five NEDs), audit committee practices (Deloitte internal audit, KPMG external audit retained annually), and Early Payment Facility accounting misclassification (£498m misclassified as trade payables rather than borrowings, per Moody's and S&P analyses).

Tertiary (flagged):

1. Wikipedia — Carillion (tertiary, used for chronology cross-check only, not load-bearing for specific factual claims).

3. OTA narrative

Observe. The information needed to read Carillion's condition accurately was produced and was in principle available to the organisation. Internal contract-review "traffic-light" systems tracked individual job margins; the finance function knew that PFI construction schemes such as Aberdeen Western Peripheral Route, Royal Liverpool and Midland Metropolitan were drifting materially behind cost and programme; the pension trustees had twice written to The Pensions Regulator (2010 and 2013) flagging deficit concerns; short-selling and credit analysts publicly identified reverse-factoring, goodwill levels and working-capital quality as red flags well before July 2017. The parliamentary joint committee found that accounts were "systematically manipulated to make optimistic assessments of revenue, in defiance of internal controls". The apparatus produced the signal; the signal did not reach the strategic-decision layer as a binding constraint. Observe was not a root cause in the strict sense — the data existed and was within reach — but it was a transmission step that carried a partly-degraded picture into the reasoning layer. The observation task was easy for the Archetype peer group of large listed UK contractors; competitors with more conservative contract-recognition policies performed it routinely.

Think. Think is the primary root-cause phase in this episode. The directors' interpretive frame — that aggressive bidding, optimistic percentage-of-completion revenue recognition, reverse-factoring-extended payables, continued dividend growth and debt-funded acquisition could be sustained through any downturn in contract performance — was wrong, and wrong on the easy end of the difficulty axis. The correct framework was available and in use at peer contractors: conservative margin recognition, transparent disclosure of supply-chain finance as debt, matching of dividend policy to free cash flow net of pension contributions, and realistic goodwill testing. The House of Commons joint committee characterised the business model as "a relentless dash for cash" and named Richard Adam as "the architect of Carillion's aggressive accounting policies"; the FCA subsequently fined three directors for misleading investor statements; KPMG's audit of the same reasoning received the largest fine the FRC has imposed, with tribunal findings of dishonesty on associated AQR submissions. The reasoning failure was therefore an Easy-Wrong Think: the correct framework existed, was accessible, and was not applied.

Act. Execution of the flawed strategy was procedurally competent and unusually persistent: Carillion continued to win public-sector work, consummate acquisitions, grow dividends and service debt for a decade. Act then compounded the Think failure at two specific decision points rather than causing it: the mid-2016 decision to pay a record £79m dividend and large executive bonuses while the pension buyout deficit was already approaching £2bn, and the post-July-2017 attempts to restructure and secure government support without a credible plan for the contract book. Act was not the root cause; by the time execution ran on the dividend, pension and rescue decisions, the reasoning that sized contract margins, treated reverse factoring as non-debt and read the acquired pension liabilities as manageable was already wrong. Act functioned as a transmission step that carried the Think failure to its external consequences; on the dividend-and-bonus-while-deficit-widening decision specifically, there is a plausible reading of Almost-wrong Act at the easy end, but the operative failure lives in Think.

4. Modality evidence

Direction. The strategic direction that set Carillion's trajectory was a specific, dated, attributable choice to grow through debt-financed acquisition and to sustain a dividend-growth record regardless of the underlying cash position of the contract book. The acquisition programme — Mowlem (£350m, 2006), Alfred McAlpine (£572m, 2008), Eaga (£298m, 2011) — was approved by the board in three discrete steps, each absorbing inherited pension liabilities and goodwill that the finance function was required to justify rather than question (HC 769; House of Commons Library CBP-8206). The operating model that accompanied this acquisition strategy combined thin-margin construction — including PFI hospital bids at or near zero profit — with a commercial posture of winning work at scale and extracting value through Extended Early Payment Facility terms imposed on the supply chain (HC 769; ACCA, Carillion's collapse and the future of PFI). The board's choice to maintain sixteen consecutive years of dividend growth, including a record payment in 2016 while the pension buyout deficit was approaching £2bn, was a directional commitment that subordinated capital allocation discipline to equity-market optics; it was approved by an identifiable board led by Chairman Philip Green (from May 2014) and executed by Chief Executive Richard Howson (HC 769; Duke FinReg Blog governance case study; FCA Final Notice — Richard Howson, January 2026). The Direction evidence meets the specificity, timing, and attribution prongs of the Direction Evidence Rule: discrete acquisition decisions, datable to specific years, attributable to the board and named executives.

The directional choice was also implicitly a choice about which risks not to manage. Bidding on complex PFI construction schemes with no margin buffer, while simultaneously absorbing multi-billion pension deficits and financing growth through reverse factoring rather than equity, was a directional posture that required each downstream component — contract execution, working capital, pension contributions, and dividend policy — to go right simultaneously. The House of Commons joint committee characterised the model as "a relentless dash for cash" and identified it as the structural premise on which the subsequent accounting and reporting failures rested (HC 769, §§ on business model and directors' responsibilities).

Structure. The governance architecture placed the audit function — both internal and external — in a structural configuration that systematically prevented independent challenge from reaching the board. The board's audit committee retained KPMG as external auditor for the full period of the episode (KPMG conducted the Carillion audit from 1999 to 2017 without break) and outsourced internal audit to Deloitte, paying Deloitte approximately £10m over the period for risk and control services (HC 769; Duke FinReg Blog). The House Committee found that Deloitte's internal audit work was "narrowly focused, did not provide the audit committee a complete picture of the control environment, and appeared not to be focused on some of the most significant risks to the company"; the committee characterised Deloitte as "either unable to identify effectively to the board the risks associated with their business practices, unwilling to do so, or too readily ignored them" (HC 769). This is a structural finding — both the internal and external audit arrangements, as configured, were organisationally incapable of surfacing contract-book risk to a level that would have constrained board decision-making.

The reporting lines for contract-margin recognition also concentrated authority in the finance function without independent counter-weight. Richard Adam served as Finance Director from April 2007 to December 2016 and held responsibility for the procedures, systems, and controls relating to financial reporting; the FCA found that those systems were "not sufficient to ensure that contract accounting judgments made in its UK construction business were made, recorded and reported appropriately" (FCA Final Notice — Richard Adam, January 2026). Authority to override optimistic margin recognition sat with Adam personally, and the organisational path from individual contract managers flagging cost overruns to a binding board-level revision of revenue recognition was not reliably functional. The audit committee, under this structural arrangement, reviewed goodwill impairment and KPMG's external audit opinion and concluded no impairment to £1.57bn of goodwill was necessary — a conclusion the 2022 FRC tribunal found KPMG had supported with dishonest conduct in audit quality reviews (FRC Sanctions, 2023; FRC Tribunal Report, 2022).

Processes. The operational machinery that should have converted internal contract-review signals into strategic decision inputs did not function as designed. Carillion's internal contract-review system tracked individual job performance through a traffic-light framework; the parliamentary joint committee found that "accounts were systematically manipulated to make optimistic assessments of revenue, in defiance of internal controls" (HC 769). The 2017 audit committee discovered eighteen contracts suffering losses — a finding that by that point had been visible in components of the internal tracking system but had not been escalated into the group financial statements or into board-level risk assessment in a form that would have triggered corrective action (HC 769; FCA Final Notice — Richard Howson, January 2026, finding that Howson was aware of serious financial troubles but "failed to respond appropriately to the warning signs").

The reverse-factoring process is a second, operationally specific process failure. Carillion imposed standard payment terms of 120 days on suppliers, directing them into the Early Payment Facility operated through Santander; this allowed Carillion to extend payables while receiving supplier-discounted early payments, but created a financial liability to the bank that the accounting procedures classified as trade payables rather than borrowings. Moody's and Standard & Poor's each identified the misclassification, estimating up to £498m of debt was presented as other creditors rather than as financial indebtedness (Duke FinReg Blog; ACCA, Carillion's collapse and the future of PFI; HC 769). The process failure is not only that the EPF was used — it is that the classification procedure, the disclosure review process, and the audit testing process each failed to correct the presentation before the accounts were signed. Santander's withdrawal of the EPF in December 2017, once Carillion's financial position became visible, immediately exposed the working-capital gap that the accounting procedure had concealed.

Scoring note (zero-modality rationale): the Processes contribution described in this subsection is classified at the boundary with Capability per the methodology §3 Processes / Capability replacement test ("if the current operating staff were replaced by new hires of comparable background, would the operational pattern survive?"). The §4 evidence applies the test explicitly and concludes that the strategic weight sits on the Capability side — the operational edge depends on the specific individuals and tacit judgement carrying it, not on documented routine. The Processes component is acknowledged in narrative but does not carry standalone weight; both modalities are evidenced and the boundary call is recorded in the audit trail. Categorisation under METHODOLOGY-ota-scoring-v4.md §5: classification boundary with an adjacent modality.

Capability. The capability dimension of the Carillion failure is narrower and more specific than the Direction and Culture failures, and the evidence base is correspondingly thinner. Carillion employed qualified engineers, project managers, and facilities professionals across its contract book; the group's size — 43,000 employees, 420 public-sector contracts — attests to operational competence at the delivery level. The capability gap the episode surfaces is at the executive and board level: the capacity to evaluate and price complex PFI construction risk, and to maintain that discipline under competitive bidding pressure. The Royal Liverpool Hospital contract was awarded after Carillion undercut its nearest competitor by £36m NPV, having initially ranked third of five bidders (ACCA; HC 769). Bidding a 0% profit margin on the Battersea Power Station redevelopment phase was recorded in committee evidence as an explicit corporate decision, not a modelling error (HC 769). The pattern across multiple large contracts suggests that the capability to assess and walk away from mispriced risk — a standard competence among peer-group UK contractors operating at comparable scale — was either absent at the executive level or was overridden by a revenue-growth imperative. The evidence does not conclusively separate a capability gap from a culture-driven suppression of risk assessment; raters should note this as a thin-evidence area requiring interpretive care.

Scoring note (zero-modality rationale): the capability gap evidenced in this subsection is real but narrow and is classified at the boundary with Culture — the §4 evidence locates the operative deficit not in technical or professional skill but in the behavioural defaults that shaped how skill was deployed (cf. methodology §3 Capability / Culture boundary test). The weight is therefore carried by Culture rather than by Capability. Categorisation under METHODOLOGY-ota-scoring-v4.md §5: classification boundary with an adjacent modality.

Culture. The cultural layer in the Carillion failure is the most extensively documented by primary sources and operates across multiple behavioural patterns. The House of Commons joint committee characterised the board as presiding over a "rotten corporate culture" and found that the board was "responsible and culpable for the company's failure" (HC 769). The mechanisms through which culture operated were multiple and reinforcing. First, norms around financial disclosure: Richard Adam, named by the committee as "the architect of Carillion's aggressive accounting policies", served as Finance Director for nearly a decade establishing a house style for contract revenue recognition that materially overstated revenues and profits and failed to comply with IAS 11 standards (HC 769; FCA Final Notice — Richard Adam, January 2026). The FCA found that Adam acted recklessly; the committee's naming of a single individual as "architect" of the policy signals that the culture had a specific originating actor and that the norm was transmitted downward rather than arising from distributed pressure.

Second, norms around internal challenge: the pattern across the audit committee's engagement with KPMG, Deloitte's internal audit, and the pension trustees' escalations (2010 and 2013 letters to The Pensions Regulator, both responded to without substantive change) is consistent with a board and executive culture in which unfavourable information was routinely absorbed without generating corrective action (The Pensions Regulator, Regulatory Intervention Report; HC 769). The relaxation of executive bonus clawback conditions in 2016 — at the point when the pension deficit was approaching £2bn — is a recorded governance action that signals the cultural priority ranking within the board (HC 769). Third, norms around external disclosure: the FCA findings against all three named directors — Howson (£237,700 fine), Adam (£232,800 fine), and Khan (£138,900 fine) — rest on recklessness findings regarding misleading investor announcements, establishing that the cultural norm around external communication was to project confidence rather than to surface deterioration (FCA Final Notices, January 2026). The KPMG tribunal's dishonesty findings on AQR submissions — fabricated audit working papers and meeting minutes — represent an external professional firm adopting the same norm in its dealings with the regulator (FRC Sanctions, 2023; FRC Tribunal, 2022). Culture is therefore load-bearing not only within Carillion but in its extended professional ecosystem.

---